A single subanesthetic dose infusion of the N-methyl-D-aspartate (NMDA) receptor antagonist ketamine has rapid and robust antidepressant effects in patients with treatment-refractory major depressive disorder (TRD). A family history of an alcohol use disorder (Family History Positive, FHP) is one of the strongest identified predictors of an improved antidepressant response to ketamine. Like ketamine, alcohol is a functional NMDA receptor antagonist. FHP is associated with differential response to both alcohol, e.g. decreased body sway and plasma cortisol, and ketamine, e.g. blunted psychotomimetic side effects. One of the primary mechanistic hypotheses for ketamine's antidepressant action is the acute intrasynaptic release of glutamate from major output neurons, e.g. cortical pyramidal cells. Preliminary clinical studies have demonstrated this acute glutamate surge in response to subanesthetic dose ketamine. Based on these findings, we hypothesize that ketamine's enhanced antidepressant efficacy in FHP TRD subjects is, at least in part, attributable to an in- creased glutamate release relative to TRD subjects without a family history of alcohol use disorder (Family History Negative, FHN). We also hypothesize that alcohol similarly augments glutamate release in this bio- logically-enriched subgroup, which may be a more objective biomarker than family history status. To test these hypotheses, we have designed a single-site, open-label inpatient study of 21-65 year old medically and neurologically healthy, currently moderately-to-severely depressed TRD patients. We plan to recruit 25 FHP and 25 FHN TRD subjects. All subjects must be free of a lifetime substance use disorder (except nicotine or caffeine) and social drinkers. The experimental portion consists of two phases. The preliminary first phase is a medication taper (if needed) and psychotropic medication-free period. The experimental second phase comprises two pharmacokinetically-defined basal-bolus alcohol and one subanesthetic dose (0.5mg/kg x 40 minute) ketamine infusions. The first alcohol infusion will establish the pharmacokinetic profile for a subsequent alcohol infusion occurring during 7T-[1H]-magnetic resonance spectroscopy (MRS) with a 2x2x2 cm voxel placed in the ventromedial prefrontal cortex/ventral anterior cingulate cortex (vmPFC/vACC). The ketamine infusion will also occur during 7T-[1H]-MRS. The primary outcome measure is group mean change in Montgomery-sberg Depression Rating Scale (MADRS) score from the pre-ketamine infusion baseline-to-one week post-infusion, where we observed ketamine's greatest antidepressant effect in FHP TRD. Additional outcome measures are vmPFC/vACC glutamate change in response to ketamine and alcohol challenge based on family history status. In summary, this study will provide key mechanistic information on ketamine's improved antidepressant efficacy in a biologically-enriched subgroup to systematically develop more efficacious, personalized treatments for major depression and reduce its enormous public health burden.